New Approaches in Time Series Analysis: Health Data Application

Abstract

It is important to analyze and interpret health data correctly. In particular, data analysis methods can make future predictions with statistical methods while also profiling the data. Estimation of the number of patients, especially in the field of health, is important for the hospital to provide quality health care to the patient and to ensure patient and health personnel satisfaction.

In this study, a monthly data set between 2010 and 2021 was simulated over the number of patients admitted to the emergency department of a state hospital in İzmir between 2020-2021 and the number of patients was estimated by time series methods. The exponential smoothing method, which is frequently used in the literature, and the Ata methods, which is a new approach, were used for modeling. Performance comparisons were made according to the mean absolute percentage error (MAPE), mean absolute scaled error (MASE) and mean absolute error (MAE) criteria, and it was determined that Ata(1,0,1)(A,N,A) method, which is a new approach in the time series, gave better results at the end of the study.